中国股票、基金及债券市场间非对称相依趋势分析——基于平滑转换SHR-Copula模型

Asymmetric Dependence Trend of Chinese Stock,Fund and Bond Market:Based on the Smooth Transition SHR-Copula Model

本文引入非椭圆SHR-Copula函数构建了Copula-GARCH模型,并将多阶段平滑转换模型应用到Copula参数的动态化中,来研究我国股票、基金和债券市场间相依关系的非对称变化。实证结果表明:三阶段平滑转换Copula模型足以刻画三个证券市场间相依关系的动态演化过程;股票、基金和债券市场两两之间的上、下尾部相依关系大体呈增长趋势,但是发生结构性突变的时点有所不同。近年来债券与股票、债券与基金之间的尾部相依性呈左强右弱的趋势,表现出显著的非对称性;股票市场与基金市场之间的上尾和下尾相依性在样本后期趋于一致,非对称情况不明显。

This paper introduces the non-elliptic SHR-Copula, develops Copula GARCH model and applied the multiple-regime smooth transition model to make Copula parameters dynamically, to research the asymmetric trend of tail dependence among Chinese stock, fund and bond market. The empirical results show that： tree-regime smooth transition model is enough to depicting the dynamic evolution process of asymmetric dependence among the three secu- rities market. In addition, the upper and lower tail dependence of stock, fund and bond market had a growing trend. In recent years, the lower tail dependence of bond-fund and bond-stock was greater than the upper tail dependence, it shows a asymmetry phenomenon; however, the upper and lower the tail dependence between stock and fund market tended to be consistent at the end of sample period, it means the asymmetry is not obvious.